---
title: "Tech – Tools & craft – Notion"
description: "Notion’s engineering collection covers tools, product design, infrastructure, security, and data systems across a set of technical posts. It opens with an inside look at the hosted Notion MCP server, created in response to requests for AI agents to work directly with Notion workspaces; the path described runs from an open-source experiment to a hosted solution with custom tools for AI workflows. Other entries describe Lore’s conversion of conversations, decisions, tasks, and durable facts into Notion pages, a Spark-native lexical-search reindexer, and multi-region systems for search, analytics, AI, and event logging. The collection also reports CI guardrails for schema changes, security testing for Custom Agents, vector-search scaling by 10x with costs reduced by 90%, Spot Balancer for Spark on Kubernetes, and the architecture and data model behind Offline Mode."
---

# Tech – Tools & craft – Notion

[Notion](https://yomu.fyi/company/notion) · Aug 10, 2026

**Type:** Explainer

## Summary

Notion’s engineering collection covers tools, product design, infrastructure, security, and data systems across a set of technical posts. It opens with an inside look at the hosted Notion MCP server, created in response to requests for AI agents to work directly with Notion workspaces; the path described runs from an open-source experiment to a hosted solution with custom tools for AI workflows. Other entries describe Lore’s conversion of conversations, decisions, tasks, and durable facts into Notion pages, a Spark-native lexical-search reindexer, and multi-region systems for search, analytics, AI, and event logging. The collection also reports CI guardrails for schema changes, security testing for Custom Agents, vector-search scaling by 10x with costs reduced by 90%, Spot Balancer for Spark on Kubernetes, and the architecture and data model behind Offline Mode.

## Context

Notion received requests for an MCP server that would allow AI agents to work directly with Notion workspaces. The collection also addresses technical challenges involving search indexing, regional data processing, schema changes, vector-search scale and cost, Spark reliability, and offline access.

## Approach / What changed

For the MCP server, Notion moved from an open-source experiment to a hosted solution with custom tools designed for AI workflows. Other described approaches include a Spark-native search reindexer, CI guardrails for breaking schema changes, Spot Balancer for Spark on Kubernetes, and an architecture and data model for Offline Mode.

## Takeaways

- Notion’s hosted MCP server evolved from an open-source experiment into a hosted solution with custom tools for AI workflows.
- The lexical-search reindexer replaced a manual, multi-week indexing process with a Spark-native pipeline described as faster, safer, and easier to run.
- Spot Balancer, open-sourced with AWS, reduced Spark compute costs by up to 90% while maintaining reliability.

**Tags:** [MCP](https://yomu.fyi/topic/mcp), [Notion](https://yomu.fyi/topic/notion), [Spark](https://yomu.fyi/topic/spark)

- Source: [Notion](https://www.notion.com/en-gb/blog/topic/tech)
- Source URL: https://www.notion.com/en-gb/blog/topic/tech
- Ingested by Yomu: 2026-09-09T05:19:32.527Z

[Read original post](https://www.notion.com/en-gb/blog/topic/tech)